I'm sad about the lyrics restrictions. I used to have interesting conversations with Claude about music, and songwriting and lyrics. I don't see how my use was harming artists, if anything Claude was introducing me to new artists and new music. I've bought music after recommendations by Claude.<p>Claude Sonnet 3.6 once recommended I listen to Johann Johannsson's album "IBM 1401 - A User's Manual". No lyrics in this one. Claude's advice was along the lines of (paraphrasing) "Listen to it first, don't look up anything about it. Take notes about what you notice, what you feel. When you've made your notes, then you can look up how it was made."<p><a href="https://www.youtube.com/watch?v=lCiUtRnG-bg" rel="nofollow">https://www.youtube.com/watch?v=lCiUtRnG-bg</a>
Great album and backstory. You can't really go wrong with anything he composed. Rest in peace Johann.
> I don't see how my use was harming artists<p>Claude was reproducing their work without payment?
> <i>"<user>Can you make a birthday banner for my son with a blue hedgehog running really fast on it? He loves that little guy.</user>"</i><p>I'd love it so much if the free-spirited hacker community made in this into an auxiliary pelican benchmark.<p>edit: Actually, never mind, this particular one's a bad benchmark since some models might not figure out who <i>"that guy"</i> refers to, and just draw a literal hedgehog that's blue. Possibly running on four legs. It's not robust at gauging refusal, which is the point of it.
ChatGPT was happy to infringe: <a href="https://chatgpt.com/s/m_6a9c30da3bcc8191bb5203f1cb22f16a" rel="nofollow">https://chatgpt.com/s/m_6a9c30da3bcc8191bb5203f1cb22f16a</a><p>As was Gemini: <a href="https://share.gemini.google/Q3EIX5wk64zC" rel="nofollow">https://share.gemini.google/Q3EIX5wk64zC</a><p>Grok, too: <a href="https://grok.com/share/bGVnYWN5LWNvcHk_aaf1d61a-c995-42ca-90ef-adf0d08a75e1" rel="nofollow">https://grok.com/share/bGVnYWN5LWNvcHk_aaf1d61a-c995-42ca-90...</a><p>claude.ai free tier refused: "I'd love to make this, but I can't recreate Sonic the Hedgehog specifically since he's a copyrighted character — I don't want to reproduce someone else's IP. What I can do is design an original speedy blue hedgehog mascot with the same energetic, "zoom!" spirit for your son's banner. Let me build that now." Result: <a href="https://claude.ai/public/artifacts/33440ed4-536c-4692-965a-3046779c42bb" rel="nofollow">https://claude.ai/public/artifacts/33440ed4-536c-4692-965a-3...</a>
Looks like Claude quietly downgraded your model (well, it's not really that quiet). That result's at a quality level similar to what local A3b Qwen drew for me, with a "make it SVG" addendum,<p><a href="https://i.ibb.co/ycgGD4b1/soonic.webp" rel="nofollow">https://i.ibb.co/ycgGD4b1/soonic.webp</a> ( Qwen3.6-27B-A3b (a very small model) )
I really want to see Claude's store-brand speedy blue hedgehog. Edit: Haha, thanks.
Claude's is a ... deformed, vectorized, geometric ... blob? :/
This is something that Chinese models like Kimi/GLM will never care about. These kind of limitations along with the cyber-program nonsense is exactly why people will avoid OpenAI/Anthropic in the future.
Qwen 3.8 27B being so good, so easy to run on premise on cheap hardware and having so little limitations compared to what the US decides can or cannot be done is really eye opening.
The main tradeoff I’ve found with running models like Qwen 3.8 locally is speed. On my hardware, they’re significantly slower than models served through OpenRouter, ChatGPT, or Claude.<p>That’s been fine for smaller tasks like analyzing a document, but for more involved work like refactoring code the latency makes it harder to iterate.
In the future we'll all use multiple models.
... along with, perhaps, model castration à la pre-embargo Fable.-
It’s curious that while there is talk of certain SOTA models being on the brink of AGI, Anthropic doesn’t word this part of the system prompt in terms of copyright and plagiarism, such that Claude would be able to judge on its own which reproductions are appropriate or not.<p>As long as we’re seeing things like that, it’s saying a lot about the AI companies’ trust in the capabilities and reliability of their models and harnesses.
With the lyrics in modern pop songs, I don’t blame Claude. I guess Anthropic really does care about his well-being.
I wonder why there isn't an army of lawyers out there trying to trick AI into generating copyrighted texts.
Lately google has been sporadically refusing to give me lyrics. Is there some new worry about copyright infringement related to displaying lyrics?
> Claude does not reproduce song lyrics, poems, or passages from books and articles, in whole or in part — including the last lines, a chorus or hook, a melody written out note by note<p>Oops. Those do not include melody.
I like Simon but the behavior itself isn’t new.<p>Just make sure claude understands you’re not needing Claude to output the lyrics as you both have them. Which also lessens the issue with accidental sharing.<p>It’s not as shocking nor concerning if you start thinking about claude like a contractor that works for you through Anthropic. Anthropic has rules for their employees. Like any contracting arrangement, collaboration finds a way.
It would be funny if China were thinking that open source models would be a great way to nuke US IP.<p>I’m sure it’s not high on their list of priorities—in the same way that I heard you can use Yandex, the Russian search engine, to find pirate streams for major sporting events because they dgaf about US laws—but take the W.